Motivations

From the write-up of my note describing the $F_2$ analysis of the '95 Shifted Vertex Data [1] to the publication of the results by the collaboration I have had several interactions with the author of the other parallel analysis, as well as with other ZEUS colleagues, in order to cross-check methods and results. Since several aspects of my analysis are not standard I would like to give in this note the final numbers on $F_2$ obtainable from those data with this method, on the light of the increased confidence of the data and on the method (and also after that some mistake has been corrected ...). The analysis strategy is described in detail in [1], and it is sketched here very shortly: from the observed number of events we infer the (``true'') number of events to be attributed to each $y$ and $Q^2$ cell. These values are then corrected for the radiative contribution, obtaining the born cross section for each bin. The $F_2$ values are calculated in the centre of the bin, using a Taylor's expansion up to the second order, to take into account the non constant behaviour of the cross section in the bin. Finally the uncertainties due to systematic effects are evaluated in a probabilistic way, also taking into account the correlations they introduce on the data points.